Tag

Nvidia Vera

All articles tagged with #nvidia vera

technology17 days ago•11 min saved

AMD Unveils 256-Core EPYC Venice: Zen 6 Architecture Targets Nvidia Vera

AMD has detailed its next-generation EPYC 9006 'Venice' server CPUs, built on the Zen 6 architecture. The flagship 256-core chip features 1GB of L3 cache, 16 channels of DDR5 memory, and PCIe 6.0 support. AMD claims the new platform outperforms Nvidia's Vera CPU by 2.24x in throughput, though independent verification is pending. The lineup includes density-optimized, frequency-optimized, and HPC-specific variants, with enterprise SKUs arriving later in 2027.

via servethehome.com|

#amd-epyc-venice #data-center-hardware #nvidia-vera

technology19 days ago•14 min saved

AMD Unveils 256-Core Venice EPYC: A Massive Architectural Shift to Challenge Nvidia

AMD has detailed its 6th-generation EPYC 'Venice' processors, marking its largest architectural shift since 2019. The flagship 256-core chip features 1GB of L3 cache, 16 channels of DDR5, and PCIe 6.0 support. AMD claims it outperforms Nvidia's Vera CPU by 2.24x in throughput, though independent verification is pending.

via The Register|

#amd-epyc-venice #data-center-cpus #hpc-performance

technology3 months ago•3 min saved

AMD Venice Set to Overtake NVIDIA Vera in 2027, Morgan Stanley Forecasts 6.75M Units

Morgan Stanley projects AMD's Venice EPYC (Zen 6, 2nm) could surpass NVIDIA's Vera CPUs in 2027—Venice about 6.75 million units vs. Vera's 5.75 million—driven by strong TSMC CoWoS packaging demand and Venice targeting both AI and HPC. NVIDIA remains TSMC's top customer, with CoWoS-L for GPUs and CoWoS-R for Vera shipments fueling a ~52% YoY rise in data-center revenue, while rising custom silicon efforts by AI firms threaten to tighten supply in the compute market.

via Wccftech|

#agentic-ai #amd-epyc-venice #morgan-stanley

technology4 months ago•6 min saved

NVIDIA Vera ARM CPU with Olympus Cores Posts Early Benchmark Wins for ARM Data Centers

NVIDIA’s Vera data-center CPU, built around its in-house Olympus cores, shows unprecedented ARM performance in early Linux benchmarks with 88 cores (176 threads via spatial multithreading), Armv9.2 ISA, up to 1.2TB/s memory bandwidth, PCIe Gen 6 and CXL 3.1. It carries a peak 450W TDP (roughly 50W+ under LPDDR5X usage) and benefits from solid upstream Linux support (Kernel 6.18+ on Ubuntu 24.04 base, GCC 16.1+/LLVM 21+ for Olympus). NVIDIA limited the initial tests to relevant AI/data-center workloads and is still tuning power management, so the results are promising but not fully representative of a production, enclosed-server environment yet.

via Phoronix|

#armv92 #data-center-cpu #linux-support